Password reset (expires every two months)
Legal instruments to combat racism on the internet
2.3. The legal basis for investigations and seizures
2.4. Obstacles in press and media law to holding a person responsible for racist content
2.5. Obstacles posed by data protection law
2.6. Problems of international cooperation among police and law enforcement authorities
The Internet, being characterised by the volatility and ubiquity of its contents (refer to Part I above), constitutes an important challenge to law enforcement authorities. Although the Internet is not a legal vacuum and the responsibilities of its actors are defined or definable (refer to Part III, below), the traditional instruments used by law enforcement authorities to establish such responsibility, i.e. tracing the actors and producing evidence of criminal acts, are not necessarily adapted to the specific technical features of the Internet. In what follows, we will analyse the scope of competence of national law enforcement authorities, the procedures they must respect in pursuing investigations and the limits of their investigative powers deriving from basic rights of the individual (data protection) and other constitutional guarantees. Apart from problems existing at the internal level, the lack of international co-operation in the area and the fact that certain countries have turned into safe harbours for hate-oriented speech, due to far going protection of the freedom of expression, constitute serious obstacles to the work of law enforcement authorities.
The Internet constitutes a global means of communication that allows users to access information from all over the world. A question arises as to the extent to which individual national authorities are competent for contents originating from outside the national territory.
In criminal matters, most countries subscribe to the principle of territorial jurisdiction, according to which the competence of national law enforcement authorities is limited to crimes committed on national territory. The place at which a criminal offence was committed is, in most countries, defined in two ways: either the place where the illegal act was committed, or the place where the result of the illegal act was felt. Given the latter part of the definition, most countries are theoretically competent for any illegal content on the Internet, provided that is accessible from their territory1.
According to § 9 of the German Penal Code (Strafgesetzbuch, hereinafter StGB), the location of a criminal act is defined as the place where the criminal acted or where the result of his action materialised or was intended to materialise.
Under the Criminal Code, French law is applicable to offences committed on the territory of the Republic. An offence is deemed to have been committed on the territory when at least one of its constituent elements has taken place in the sea, land or air space of the French Republic (Art. 113-2).
Article 6 of the Criminal Code lays down the principle of the territorial application of criminal law: any person who has committed an offence on Italian territory is liable to be punished according to Italian law. Article 6 paragraph 2 states that an offence is to be regarded as having been committed in Italy where the act or omission took place there, in whole or in part, or where the event resulting from that act or omission was produced there (ubiquity).
These examples show that national provisions concerning territorial jurisdiction give States a very broad jurisdictional competence. The French Conseil d’Etat described the situation as follows:
“It follows from those provisions that French criminal law clearly applies in the case of an offending message available on the Internet network, no matter where in the world its source is situated ... Such a mechanism in reality has the effect of dissociating the place where the offence was actually committed from the place where it produces its effects, and gives the national courts a very broad jurisdiction.2”
This very broad competence is considered problematic in some states, because it forces anyone who publishes on the Internet to check all the legal orders of the world if he or she wants to be sure that his publication will not constitute a criminal offence. For this reason, and because such legal uncertainty may constitute obstacles to economic development in the area of Internet services, attempts have been made in some spheres to restrict the general rules on criminal jurisdiction when these are applied to Internet.
Legal commentators propose to limit criminal jurisdiction in respect of the Internet to those acts with which the author really intends to reach a German audience3. Only contents tailored to a German audience should fall within German jurisdiction.
The proposed Directive on electronic commerce4 effectively limits jurisdiction to the country of origin of the service provider (Art. 3 of the Directive). The service provider can only be held responsible according to the legal provisions in force at the place of his business5. However, by way of derogation, the Member States may take measures for the purpose of fighting incitement to hatred on grounds of race, sex, religion or nationality (Art. 22, para. 2 of the Directive).
Jurisdiction in civil matters concerning the Internet is much less clear than criminal jurisdiction. In the fight against racism on Internet, civil jurisdiction becomes relevant where the author of illegal content can be held responsible on the basis of tort law, which means that he is either liable in damages (for example in Italy; refer to Part III, below), or the addressee of a cease and desist order. Where the access or host provider cannot be held responsible under criminal law, a civil law cease and desist order aimed at cutting access to an Internet site with racist content can be an efficient alternative means of stopping the distribution of such racist content.
As in criminal cases, a question arises as to the extent of the competence of the national court when the author and/or Internet host of the racist content is located in a foreign country6.
In the only Italian decision7 to deal with the question of the international jurisdiction of an Italian court in respect of an unlawful act committed on the Internet, the President of the court granted an application for an injunction concerning defamatory information published on an “open” Internet site, although the server was in the United States. The respondent’s argument that the Italian court lacked jurisdiction was rejected on the ground that the court was competent to deal with a complex activity involving the dissemination of information which was harmful to others, since this information, by being placed on the website, was directly accessible in all countries linked to the Internet. Accordingly, the fact that the website concerned was opened abroad and the information loaded abroad could not preclude the jurisdiction of the Italian courts.
Once the competent court is determined, there arises a more complicated question as to which law a judge must apply when deciding about the legality of content on a server located abroad. The diversity of rules existing in this regard entails the risk that a domestic judgment will not necessarily be enforced in the foreign country where the illicit content is stored on a server. Harmonisation of rules would minimise this risk.
Although, in theory, states are competent for the prosecution of Internet crimes committed anywhere on the globe, they will not succeed in holding responsible the person who published illegal content if this material was posted to the net in a country where such content is protected by law. In respect of racist content, due to a very broad conception of the freedom of speech, the United States of America has become a safe data harbour in which prosecutions in respect of racist content or the enforcement of any foreign judicial decision will be extremely difficult. The fact that a majority of racist sites are located in the US makes it necessary to explain the US-American concept of freedom of speech in more detail.
Although the United States has actively pursued a policy agenda intended to combat behavior motivated by hatred8, the fight against hate motivated communications on the Internet faces some specific obstacles. These communications are classified as speech in the American legal system and, as such, the special protection accorded to freedom of speech by the First Amendment to the United States federal Constitution9 makes the regulation of their content extremely difficult. As a result, and despite numerous attempts, there is currently no comprehensive federal legislation prohibiting racism on the Internet.
The importance of free speech has historical roots. The founders of the United States came to America to avoid persecution - often at the hands of their own governments - for their political or religious ideas. Their goals in establishing a new system of governance were to guarantee individual freedom and to place strict controls on those in power. Censorship - particularly of ideas which might be unpopular – was anathema to a truly democratic process and was therefore to be eradicated at all costs.
Even now, freedom of speech is seen as the most fundamental of rights in the United States. The theory most frequently advanced is that only through open public debate will the truth become clear. Such debate is possible only if no ideas, no matter how unpleasant, are censored. This public debate - in particular the ability to be informed about and to criticize the government as well as majority-held opinions - is an essential prerequisite for all citizens to perform their self-governing function. The specter of legal sanctions imposed on a particular type of speech might have a "chilling effect" on legitimate speech: an individual might refrain from stating his opinions for fear of punishment, thereby depriving the public of this point of view.
In order to protect the individual's rights from governmental censure, the American legal system requires judges to verify the constitutionality of any law under which an accused may be prosecuted. In order for the government to be able to regulate an area, not only must adequate legislation be drafted and enacted, such legislation must pass the judicial test of constitutionality before it can be enforced.
In particular, a law which places any restriction on the freedom of speech is subject to the strict scrutiny of the courts. The restriction must be clearly defined, justified by a compelling interest of the government and limited to what is strictly necessary10 to achieve the specific goal. Restrictions based on the content or subject matter of speech might allow the Government to effectively silence certain ideas and are therefore presumptively unconstitutional.
Only in the area of obscenity have courts tended to find that government interests in regulation outweigh the primacy of free speech11. That fact, combined with strong political pressures concerning the protection of children from cyberporn, may explain why the only pieces of legislation attempting to regulate the content of Internet communications adopted on a federal level to date have targeted indecent rather than racist communications.
The first such attempt was the Communications Decency Act of 199612 (CDA) which provided for both civil and criminal penalties for the use of an interactive computer service to knowingly transmit, send or display certain material of a sexual or excretory nature to minors. The U.S. Supreme Court held significant provisions of the CDA to be unconstitutional13 as they related to "indecent" or "patently offensive" materials on the grounds that it was not the least restrictive means to achieve the government’s goals. Other provisions, in particular those providing certain defenses to liability for service providers and those prohibiting communication of obscene e-mails intended to harass or annoy its recipient14, remain in force.
Congress subsequently adopted the Child Online Protection Act (COPA) which targets commercial websites that disseminate information "harmful to minors" without restricting underage access to such materials15. COPA represents Congress’ attempt to remedy the constitutional defects in the CDA. COPA’s constitutionality has, however, been challenged as being vague and overbroad because it threatens speech that is protected as to adults. A federal district court has granted a preliminary injunction against enforcement of COPA on the ground that COPA is presumptively invalid as a content-based regulation of non-obscene sexual expression16.
State laws specifically prohibiting racist expression have not generally survived claims of unconstitutionality. For example, in R.A.V. v. St. Paul17, the U.S. Supreme Court struck down a city’s ordinance that made it illegal to exhibit a burning cross, a swastika, or other inflammatory symbol that might arouse "anger, alarm, or resentment in others on the basis of race, color, creed, religion or gender". All nine justices agreed that such acts were reprehensible, but that the ordinance was unconstitutional. There was, however, considerable disagreement concerning the rationale supporting the judgment, and the case generated four separate published opinions.
Other laws, particularly those concerning fair housing and employment, which do not specifically target racist speech have sometimes been used successfully to prosecute or prohibit racist speech. The California Supreme Court, for example, has held that racist comments directed at an employee by his superior may even, in some circumstances, be subject to a prior restraint. Where such comments "contribute to a hostile or abusive work environment" and therefore constitute employment discrimination, such speech may be enjoined at the workplace18.
Although few state laws specifically address communications over the Internet, courts are generally willing to apply existing doctrines, such as employment and housing discrimination, to electronic messages. Two lawsuits for employment discrimination have recently been filed based on the circulation at the workplace of racist e-mails19. Although neither suit has yet reached a trial on the merits, one has survived a motion to dismiss20. The Pennsylvania Attorney General sought an injunction against a white supremacist group and all persons and entities having control of that group’s website in connection with threatening material against an individual on that site21, and federal authorities subsequently filed charges for civil rights violations22. Defamation statutes have been used when a specific individual is targeted, or in states which have enacted group libel statutes.
Other legal theories that have been used include harassment, ethnic intimidation, and mental distress of a person to whom abusive language is address or hearing abusive language addressed to another. The issue in these cases "invariably turns on the issue of whether the language in question is so outrageous that society wishes to permit recovery, and this requires an examination of the nature of the language at issue in particular cases . . . Recovery has tended to be permitted in such cases where the plaintiff has been placed in reasonable fear of physical harm or of some other clearly intimidating experience such as arrest or imprisonment."23
First Amendment law is a highly controversial area of American law, particularly as it concerns expressions of racism, and it poses a classic case of the competing imperatives of liberty and equality. The raison d'être of the First Amendment guarantee of freedom of speech is the protection from persecution of those holding a minority view. Yet it's strict application creates serious obstacles to the protection of certain minorities from verbal abuse. The development of the Internet - in particular the ability to reach an increasingly large audience while remaining totally anonymous and therefore free from editorial or societal pressures - has added a new level of complexity and urgency to the fight against racism in the United States.
This problem of “safe havens” is not limited to racist contents, but also concerns revisionist sites whose existence has to do with the fact that there is no criminal legislation in that regard in certain European countries.
The criminal or civil responsibility of a particular author of illegal content can only be established on the basis of evidence. The volatility of electronic content confronts the investigative authorities with major difficulties (refer to point 2.2., above). They are expected to react fast, whereas the legal means of intervention do not effectively permit quick intervention or seizure of evidence in most countries. The prerequisites for obtaining a search warrant are very often ill adapted to the form which illegal content takes on the Internet. Moreover, the technical means for an efficient investigation, based on tracing data transfers back to the author, are very often in the hands of the access or host provider. This poses the question of whether access providers are obliged to help police in the process of obtaining or securing relevant data files. Another limitation on tracing back illegal content, especially in the field of e-mail, is the fact that surveillance of the private communication is only possible within the narrow scope of the laws on the interception of postal and telephonic communications.
To be able to search a building, police officers require - as in most countries – a judicial search warrant (139 ff StPO). Having gained access to a building, police may seize material which is relevant to the investigation, including data files and computers. The owner of any data base is obliged to co-operate for this purpose. Co-operation can consist of production of copies of data files, decryption of encrypted messages or processing data in such a way that it can be used by the law enforcement authorities. Moreover, according to § 89 of the Telecommunications Act, the operators of telecommunication services are obliged to provide to the police all the installations which are necessary for the surveillance of telecommunications. On the other hand, access providers, other than public institutions, who know of the existence of illegal content, are not obliged to bring it to the attention of the authorities (§§ 84 and 86 of the Criminal Procedure Act).
Art. 5 of the Telecommunications Data Protection Directive requires that interception or monitoring of communications, on both public and non-public networks, take place only when legally authorised.
As in other countries, police may seize materials which are linked to the commission of a criminal offence. However, they may not seize e-mails which are held in intermediate storage on the server of an access provider. The storing of e-mails is still part of the communication process, so that seizure would have to be qualified as an interception of telecommunications. Under a constitutional guarantee (Art. 10 Grundgesetz) such an interception of telecommunications can only be effected by the Federal Office for the Protection of the Constitution (Bundesamt für Verfassungsschutz) or by the corresponding institutions of the Länder. In a case decided by the Regional Court of Hanau24, a public prosecutor was forbidden to seize incoming and outgoing e-mails which were still stored on the server of the access provider. The prosecutor had ordered their confiscation because he suspected orders for pornographic materials to be among them.
Concerning certain extremely serious criminal offences, Internet providers are obliged to inform the authorities of the planning or commission of such an offence (§ 138 StGB). The offence of incitement to racist hatred, however, does not rank among them.
The main legislation which permits British police to search for evidence of an offence committed by use of the Internet is the Police and Criminal Evidence Act 1984 (Chapter 60 of 1984). Under Part II of that Act, police are normally required to obtain a warrant from a Justice of the Peace if they wish to enter premises in order to search for and seize evidence of an offence having been previously committed. A first prerequisite is that there be "(…) reasonable grounds for believing" that a "(…) serious arrestable offence" had been committed (subpara. 8(1)(a)). To use words or display written material that is intended or likely to stir up racial hatred is an arrestable offence under subsec. 18(3) of the Public Order Act 1986. Whether the offence is "serious" must be evaluated by the judge on each application. Secondly, the police must show either that they would be unable to obtain access to the premises or the evidence without the warrant (subparas. 8(3)(a)-(c)), or that asking for access would probably result in the destruction or disappearance of the evidence (subpara. 8(3)(d)). One imagines that this will often be the case with respect to racist material, particularly if it is stored on computers for distribution over the Internet. Para. 5 of Schedule 1 to the Act provides that, in cases where the material sought is "contained in a computer", then the search warrant shall be interpreted as an order obliging the person in possession of the material to produce it in a visible and legible form in which it can be taken away by the police. Thirdly, the police must describe, as closely as is practicably possible, the individual articles which they believe exist and which they wish to seize (subpara. 15(2)(c)). The warrant permits entry to the specified premises on only one occasion (subsec. 15(5)). In the result, if the police are to obtain authorisation to seize evidence of propogation of racist material, they must first find out by other means that a relevant offence has been committed. Hence the importance of information provided by Internet users via hotlines or other means of information. Once a warrant has been obtained, the police can seize anything on the premises searched which amounts to evidence of any offence that has been committed. Subsec. 19(4) specifically authorises the police to "require any information which is contained in a computer and is accessible from the premises to be produced in a form in which it can be taken away and in which it is visible and legible". Nevertheless, the police find these statutory constraints frustrating in the context of racism on Internet, as they do not permit the police to maintain ongoing surveillance of what material is being propagated by known racists or racist groups from time to time25.
The Public Order Act 1986 makes additional provision for the police to be able to search for and seize material which is specifically in breach of sec. 23 of that Act, namely material that is likely to stir up racial hatred and is in the possession of the accused for the purpose of being distributed, shown or played. Under subsec. 24(1), such a search operation can only be conducted upon the authority of a warrant issued by a Justice of the Peace and it is up to the police to show reasonable grounds for suspecting that a person has a racist writing or recording in his possession. There is no guidance on the interpretation of this provision, but the wording gives reason to believe that it would also be treated as restricted to allowing a search for and seizure of a particular existing document.
Particularly useful for identifying the authors of racist and other illegal material on Internet is the possibility provided by sec. 2 of the Interception of Communications Act 1985 (Chapter 56 of 1985) to covertly survey messages passing over public telecommunications systems. This requires a warrant issued by the Home Secretary [= Minister of the Interior] inter alia "for the purpose of preventing or detecting serious crime" (subpara. 2(2)(b)). The Home Secretary has power to issue such a warrant whenever "he considers that the warrant is necessary" for that purpose, in the sense that the relevant information is necessary for that purpose and that it could not "reasonably be acquired by other means" (subsec. 2(3)). Unfortunately (from the point of view of law enforcement), the information obtained from such surveillance cannot be used to bring a prosecution against any offendor: the House of Lords held in R. v. Preston et al26, that the result of subsecs. 2(2)(b) and 6(3) combined is that information so obtained must be destroyed as soon as the process of detecting crimes has been completed and therefore cannot be used to prosecute crimes. This legislation is accordingly useful for identification of the authors of racist material on Internet, but in order to prosecute the authors, the police will need to proceed, on the basis of the information so obtained, to find independent evidence of the commission of a criminal offence (Refer also to point 2.5, on data protection, below).
In a leading decision,27 the Federal Court recently held that e-mails are covered by the secrecy of telecommunications. It thus upheld the appeal of a technical intermediary (Swiss Online), which refused to disclose the identity of the author of an e-mail.
Swiss Online had refused to comply with the request of a Zurich judge who had requested it to identify the author of an e-mail who sought to extort funds from a private undertaking. This judge, who was responsible for the investigation considered that he was not required to seek the approval of the judge specially designated to lift the secrecy of telecommunications.
The Federal Court took the view that any step in the judicial procedure which sought to identify retroactively the author of an e-mail must have a legal basis and that the approval of the judge specially designated by cantonal procedure to lift the secrecy of telecommunications was essential.
The Federal Court also emphasised the need for legislation in an area which was constantly changing.
In many countries, publications on Internet fall within the scope of application of the national press laws. This classification of Internet content as press or media has important consequences for the investigation of illegal content. On the one hand, the prescription period is very often shorter than that applicable to normal criminal proceedings. On the other hand, the applicability of press laws limits the scope of preventive measures that may be taken against the operators of websites which periodically host illegal contents.
The conditions under which Internet publications will be considered a product of the press are divergent:
According to a judgement of the Oberlandesgericht Wien of 26 November 1997, the Internet is to be counted among the forms of media in the sense of the Act on Media (Mediengesetz, the Austrian press law). This decision was subsequently criticised for generally qualifying Internet as a "medium" without examining further requirements such as periodicity of publication or the journalistic processing of content.
In general, Internet content can be considered as a product of the press if it functionally replaces the print media28. To obtain this qualification, particular content has to be processed in a journalistic and editorial manner. The administrative court of Düsseldorf29 held that a content provider who gathers information and provides a forum to third parties for their publications does not benefit from the privileges of press law, because the mere fact that the content provider structures and administers the information contained on his web-page is not sufficient to constitute journalistic processing of such content.
In a case concerning the distribution of nazi symbols, the Higher Regional Court of Frankfurt30 defined the meaning of "publishing" in the context of the Internet. According to the Court, only the first publication of a specific content on the Internet is protected by press laws. If the content (in the pertinent case, a video game wherein nazi symbols were used) is re-distributed by a third person (here, the owner of a mailbox), the this redistribution cannot be considered publication within the meaning of the press laws.
Courts have implicitly recognized that publications on the Internet, provided they appear periodically, can fall within the scope of press laws31. However, most legal commentators are opposed to an assimilation of the Internet to the press32.
In addition, the consequences of such classification as products of the press vary considerably between the countries examined:
The Cour d’appel de Paris found that the press laws and in particular the short prescription period of 3 months are applicable. However, the prescription period is not to be calculated in the traditional manner:
“The civil judgment delivered on 10 July 1997 following the proceedings commenced on 8 April 1997 by the UEJF (Union of Jewish Students of France), states that Jean-Louis Costes relied in his defence, inter alia, on the fact that the texts in question had been published on the Internet network on 14 September 1996 and that, consequently, if they did constitute an infringement of the 1881 Act, they could not give rise to criminal proceedings, since the prosecution was already time-barred.
The accused seeks to take advantage of that time-limit, the principle of which is laid down in Article 65 of the Law of 29 July 1881 on the press.
The application of Article 65, which lays down the principle of a three-month period from the first day of publication beyond which the prosecution is barred, has formed the subject-matter of a consistent line of decisions in relation to writing or images disseminated on paper (books, newspapers, posters etc.) or in an audiovisual form (radio, television, cinema etc.) where it is easy to determine the first day on which the writing or image was made available to the public, if only because it is apparent from the medium itself (newspapers, audiovisual message) or because the time when it was made available to the public corresponds to a specific act (mailing in the case of books).
In order to apply Article 65, it is necessary to determine the date on which the item was first made available to the public, since the principle is thus laid down by the legislature that after three months, in derogation from the general criminal law, the prosecuting authorities and the parties seeking civil damages are no longer authorised to initiate a prosecution in respect of written works, since the disruption of public order deemed to result therefrom or the harm caused to third parties must be regarded as lapsed or made good.
In such a situation the publication results from the renewed intention of the person who places the message on a site and chooses to keep it there or to remove it when he deems it appropriate. The act of publication is therefore continuous. The fact that the offence continues to be committed throughout the relevant period is a concept of positive law in criminal matters and applies to the definition of a number of offences.
The Court therefore finds that by choosing to keep the texts in issue available on his site on the dates on which it has been established that they were there, and in this instance on 10 July 1997, Jean-Louis Costes published them again on that date and ran the risk that the three-month period would begin to run anew from that date”.33
In a recent decision of the First Instance Court of Brussels dated 2 March 200034, the Belgian judges followed the French approach in the Costes case. They rejected the argument of the defendant, against whom a preliminary injunction was being sought in connection with a defamation action, that claims arising out of the relevant publication were prescribed. The Court found that "un délit de presse sur Internet doit être considéré comme un délit continu, tant que le texte litigieux reste aisément accessible à toute personne naviguant sur le net à la recherche d’information sur un sujet donné".
The qualification of Internet content as a press publications would limit police intervention before the publication of an article. This rules is to avoid censorship. The police must wait until content is published before intervening. The seizure of products of the press can only be effected on the basis of a judicial warrant following publication (§§ 97 and 98 StPO). There is no possibility of intervening before publication.
In investigating and tracing racist content on the Internet, the law enforcement authorities depend on the access providers’ willingness to co-operate. However, their co-operation is very often limited by the laws on data protection. In most States, such laws are rooted in the fundamental right of freedom from interference with private life and therefore constitute a pivotal counter-weight to police investigational efforts.
Both, the Federal Act on Data Protection and Telecommunications Services (Teledienstedatenschutzgesetz35) and the Compact of the Länder on Media Services (Mediendienststaatsvertrag) contain strict provisions on data protection applicable to Internet which go beyond the general data protection rules and aim particularly at guaranteeing the right to anonymity.
According to these provisions, the use and the collection of fees must be effected anonymously or by using a pseudonym. The establishment of a user profile can only be linked to a pseudonym which makes it impossible to link the user profile to data that would allow the identification of a person. Moreover, the service provider has to guarantee that the client can use the services without being identified by third parties. This provision implements the constitutional guarantee of the inviolability of communications. Data concerning the manner in which clients use the services must be deleted immediately after use, unless longer storage is necessary for billing purposes. Data on the use of Internet services may only be stored, processed or used if this is necessary for modifications of the contractual relations between the service provider and the user, if they are necessary to allow the user to use the Internet, or for billing purposes. In general and subject to narrow exemptions, data obtained for the above mentioned purposes may not be forwarded to third parties unless the user agrees. The Federal Office for the Protection of the Constitution is pressing for the enactment of a provision which would allow data to be passed to the competent law enforcement authorities36. However, this proposal has not yet been carried out.
The Federal Data-Protection Officer has no objection to an obligation to keep data on persons who are not suspected of any offence, as a precaution, but considers that keeping such data is a serious breach of personal rights which must be approved by Parliament, in other words it cannot merely be ordered by the Executive.37
Under both the Data Protection Act 1984 and the revised Data Protection Act 1998, ISPs are generally obliged to respect the confidentiality of "personal data" concerning their customers. That expression would include not only the contents of their e-mail and chat correspondence, but also details of the websites which they visited, the newsgroups to which they posted and so on. An ISP who amasses such information and discloses it to any third party without legal authority is liable to criminal prosecution. Of course, the Acts contain exemptions for the benefit of police operations. Subsec. 28(3) of the 1984 Act and subsec. 29(3) of the 1998 Act both exempt personal data from the normal non-disclosure requirements in so far as the disclosure is made for the purpose of prevention or detection of crime or apprehension or prosecution of offenders and non-disclosure "would be likely to prejudice any of [those] matters ..."
In the result, the Data Protection Acts authorise ISPs to disclose information about their customers to the police where there is a reasonable suspicion of criminal activity, but do not force them to do so. An agreement has been reached between ISPs and the Association of Chief Police Officers, according to which the police will prepare notices in a certain form and containing certain information whenever they wish to obtain personal data concerning the customers of ISPs, but have not obtained warrants authorising them to demand that information. However, ISPs generally respond positively to such notices only in so far as the personal data sought is felt to be of a "non-sensitive" nature38.
Sovereign acts of law enforcement authorities can only take place within the national territory. Hence, law enforcement authorities depend on the cooperation of foreign authorities when they want to investigate foreign authors. We know of no attempts by European law enforcement authorities to attack racist material hosted on servers located outside the national territory, other than by asking their own national providers to block access to these sites. International police investigations within Europe are often effected via Interpol or Europol. Although a highly functional forum for police cooperation, Europol is not vested with any particular mandate in respect of Internet crimes. No specific police cooperation modules have been set up exclusively for the Internet. The Council of Europe addressed this problem in its Recommendation No. R (95) 13 concerning problems of criminal procedural law connected with information technology of 11 September 1995:
"17. The power to extend a search to other computer systems should also be applicable when the system is located in a foreign jurisdiction, provided that immediate action is required. In order to avoid possible violations of state sovereignity or international law, an unambiguous legal basis for such extended search and seizure should be established. Therefore, there is an urgent need for negotiating international agreements as to how, when and to what extent such search and seizure should be permitted.
18. Expedited and adequate procedures as well as a system of liaison should be available according to which the investigating authorities may request the foreign authorities to promptly collect evidence. For that purpose the requested authorities should be authorised to search a computer system and seize data with a view to its subsequent transfer. The requested authorities should also be authorised to provide trafficking data related to a specific telecommunication, intercept a specific telecommunication or identify its source. For that purpose, the existing mutual legal assistance instruments need to be supplemented."
In a certain way, the European Union has responded to those needs in adopting its "Legislative resolution embodying Parliament’s opinion on the draft Council Resolution on the lawful interception of telecommunications in relation to new technologies39". In this resolution, the Council lists the "requirements" to be met in the Member States for the lawful interception of new telecommunications technologies, namely the Internet. The resolution aims at creating common standards for interception, thereby making cooperation between national police forces easier and less bureaucratic.
1. See also Report of the Council of Statet, “Internet et les réseaux numériques”, 1998, p. 167, http://www.internet.gouv.fr/francais/textesref/rapce98/accueil.htm
2. Report of the Council of State, cited in footnote 8 above, p. 167.
3. See Collardin, Straftaten im Internet, Computer und Recht 1995, p. 618 (621).
4. To be found on: http://www.ispo.cec.be/Ecommerce/legal/legal.html.
5. More details: Spindler, Multimedia und Recht 1999, 199, 206.
6. According to the Brussels Convention of 1968 and the Lugano Convention of 1988, of which most of the European States are members, the victim can choose between a court in the country of the defendant's domicile or a court in the country where the damage was caused or a court in the country where the damage arose. Of course the conventions only apply when the defendant is domiciled in one of the States Parties to the conventions.
7. Order of the President of the Teramo District Court of 11 December 1997 (civil matter, action for an injunction, case of defamation), in Diritto dell’informazione e dell’informatica 1998, p. 370.
8. Heather De Santis, Combatting Hate on the Internet: An International Comparative Review of Policy Approaches, Strategic Research and Analysis SRA-350, Department of Canadian Heritage, 1998, p.31.
9. “Congress shall make no law respecting an establishment of religion, or prohibiting the free exercise thereof; or abridging the freedom of speech, or of the press; or the right of the people peaceably to assemble, and to petition the Government for a redress of grievances." (italics added) USCA Const. Amend. I.
10. Supreme court opinions have generally referred to this element as requiring that legislation be “narrowly tailored" (as opposed to being “overbroad") and that it represent the “least restrictive means" for achieving the government’s goal.
11. John F. McGuire, When Speech is Heard Around the World: Internet Content Regulation in the United States and Germany, 74 NYULR 750, 753 (1999).
12. Publ. L. N° 104-104, 110 Stat 133 (codified in scattered sections of 47 U.S.C.)
13. Reno v. ACLU, 117 S. Ct. 2329 (1997) (invalidating portions of the Communications Decency Act of 1996 as not representing the most restrictive means to achieve the government's goals).
14. ApolloMedia Corp. v. Reno, 1998 WL 853216 (U.S. (Cal.) 1999) amended by 119 S. Ct. 1459 (US 1999) as cited in Andrews Computer & Online Industry Litigation Reporter “High Court Upholds Law Banning ‘harassing’ or ‘annoying’ e-mail" May 18, 1999.
15. COPA provides for fines and/or imprisonment for whoever knowingly makes a communication for commercial purposes by means of the World Wide Web that is available to any minor that includes any material harmful to minors. Publ L. 105-277 112 Stat. 2681-736 codified as 47 U.S.C. §231.
16. ACLU v. Reno, 31 F.Supp2d 473 (E.D. Penna. 1999)
17. 505 U.S. 377 (1992).
18. Aguilar v. Avis Rent A Car System, Inc. 87 Ca. Rptr.2d 132 (Ca. S.Ct. 1999).
19. Owens v. Morgan Stanley & Co., 1997 WL 403454 ( S.D.N.Y. 1997); Curtis v. Citibank, 1998 WL 3354 (S.D.N.Y. 1998).
20. Owens v. Morgan Stanley & Co., 1997 WL 793004 (S.D.N.Y. 1997).
21. “AG’S Complaint Says Hate Group Published Terroristic Threats on the Internet", The Legal Intelligencer Vo.219, N° 79, October 21, 1998
22. “Feds Target Web Threats", The National Law Journal Vol 22, N° 23, January 31, 2000.
23. “Civil liability for Insulting or Abusive Language – Modern Status", American Law Reports ALR 4th, Vol. 20 (1983) Current through the September, 1999 Supplement.
24. LG Hanau, Beschluss vom 23.9.1999, NJW 1999, S. 3647.
25. Source: Detective Chief Superintendent Keith Akerman, Hampshire Constabulary, Chairman of the Computer Crime Working Group of the British Association of Chief Police Officers.
26.  2 Law Reports, Appeal Cases 130
27. Deliberations of 5 April 2000 in Case 1A.104/1999.
28. Vid. Martin Bulliger, Ordnung oder Freiheit der Multimediadienste, JZ 1996, p. 385, 387.
29. VG Düsseldorf, Beschluss vom 25.6.1998, NJW 1999, p. 1987.
30. OLG Frankfurt a.M., Urteil vom 18.3.1998, NSTZ 1999, p. 356.
31. Rome District Court, 6 November 1997.
32. V. Zeno-Zencovich, La pretesa estensione alla telematica del regime della stampa: note critiche, in: Diritto dell’informazione e dell’informatica 1998, p. 15 ss; P. Costanzo, Libertà di manifestazione del pensiero e "pubblicazione" in Internet, ibidem, p. 372 ss; M. Franzoni, La responsabilità del provider, in AIDA 1997, p. 150.
33. Paris Court of Appeal, judgment of 15 December 1999, published on http://www.legalis.net/jnet/
34. Brussels District court, No 2000/77/C on the list of urgent applications, summary and further links available on http://www.juriscom.net
35. Act of 22 July 1997, BGBl. I, S. 1872
36. Cf. Bericht des Bundesamts für Verfassungsschutz, http://verfassungsschutz.de/publikationen/gesamt/exint06.htm.
37. Report of the activities of the Federal Officer, 1998, p. 236.
38. Source: Detective Chief Superintendent Keith Akerman of the Hampshire Constabulary, Chairman of the Computer Crime Working Group of the British Association of Chief Police Officers.
39. Official Journal C 279 of 1 October 1999, p. 498.